Hacker News new | ask | show | jobs
by silverlake 435 days ago
I think BF16 and FP16 are 1979 TFPOPs, but FP8 is 2x faster at 3958 TFLOPs. So only 10% efficiency, down from 20%. That’s not good.
1 comments

That’s with sparsity. So it’s 29% down from 40%.